Papers › ViTAEv2: Vision Transformer Advanced by Exploring Inductive Bias for Image Recognition...

ViTAEv2: Vision Transformer Advanced by Exploring Inductive Bias for Image Recognition and Beyond

21 Feb 2022arXiv:2202.10108archive 2025-07-28

Qiming Zhang, Yufei Xu, Jing Zhang, DaCheng Tao

Vision transformers have shown great potential in various computer vision tasks owing to their strong capability to model long-range dependency using the self-attention mechanism. Nevertheless, they treat an image as a 1D sequence of visual tokens, lacking an intrinsic inductive bias (IB) in modeling local visual structures and dealing with scale variance, which is instead learned implicitly from large-scale training data with longer training schedules. In this paper, we propose a Vision Transformer Advanced by Exploring intrinsic IB from convolutions, i.e., ViTAE. Technically, ViTAE has several spatial pyramid reduction modules to downsample and embed the input image into tokens with rich multi-scale context using multiple convolutions with different dilation rates. In this way, it acquires an intrinsic scale invariance IB and can learn robust feature representation for objects at various scales. Moreover, in each transformer layer, ViTAE has a convolution block parallel to the multi-head self-attention module, whose features are fused and fed into the feed-forward network. Consequently, it has the intrinsic locality IB and is able to learn local features and global dependencies collaboratively. The proposed two kinds of cells are stacked in both isotropic and multi-stage manners to formulate two families of ViTAE models, i.e., the vanilla ViTAE and ViTAEv2. Experiments on the ImageNet dataset as well as downstream tasks on the MS COCO, ADE20K, and AP10K datasets validate the superiority of our models over the baseline transformer models and concurrent works. Besides, we scale up our ViTAE model to 644M parameters and obtain the state-of-the-art classification performance, i.e., 88.5% Top-1 classification accuracy on ImageNet validation set and the best 91.2% Top-1 accuracy on ImageNet real validation set, without using extra private data.

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

The archive's official repository (ViTAE-Transformer/ViTAE-Transformer) was not harvested; the 25 samples below come from 3 community repositories.

Syntology 25 code samples harvested from 3 repositories linked to this paper: 22 ran (of which 8 constructed an object rather than computing a result; 16 with no instrument failure: 0 honoured, 0 violated, 16 with no contract checked; 6 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 4 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 23 samples from 2 repositories, 20 ran (16 with no instrument failure; 4 where Syntology's instrument failed); community: 2 samples from 1 repository, 2 ran (0 with no instrument failure; 2 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 2 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 5 of the 22 samples that ran have no recorded date.

ViTAE-Transformer/ViTAE-Transformer officialmentioned in papermentioned on GitHubpytorchSyntology: not harvested report
ViTAE-Transformer/ViTDet mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Exploring Plain Vision Transformer Backbones for Object Detection (arXiv:2203.16527; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
tamer-saleh/s1gflood-detection mentioned on GitHubpytorchCC0-1.0Syntology: 11 samples, 11 ran (4 instrument-failed) · 0 unverified report
vitae-transformer/vitae-transformer-remote-sensing mentioned on GitHubSyntology: not harvested report
MS-P3/code7 mindsporeSyntology: not harvested report
pwc-1/Paper-9 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to RandAugment: Practical automated data augmentation with a reduced search space (arXiv:1909.13719; that paper's own run record: 58 ran (of which 1 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 1 violated, 6 with no contract checked; 51 where Syntology's instrument failed) · 7 unverified); nothing checks that they implement this paper's method. report
yangyucheng000/Paper-2 mindsporeSyntology: not harvested report
yangyucheng000/papercode-2 mindsporeSyntology: 12 samples, 9 ran (0 instrument-failed) · 3 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

25 samples harvested: 22 ran (of which 8 constructed an object rather than computing a result; 16 with no instrument failure: 0 honoured, 0 violated, 16 with no contract checked; 6 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

4ran · our draft was wrong
2ran; the contract check could not run
16ran
3unverified

Licence: 4 of the 25 samples are pointer only, meaning Syntology does not serve that copy's text. For 2 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 3 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. For 7 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

tamer-saleh/s1gflood-detection community (archive-listed) 11 samples, 11 ran (4 instrument-failed) · 0 unverified
window_partition in 377 papers
census: 13 implementations from 17 papers, 5 distinct outputs across 3 buckets, one shared input per bucket
swin.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted CC0-1.0 (permissive) · 144d10b49baeb8a6 · report fixture: signature x:Tensor/4/float, window_size:int/0/int · returned Tensor · level recorded, date not recorded
ViTAE_Window_NoShift_12_basic_stages4_14 models.py ran CC0-1.0 (permissive) · 1b9001750c072436 · report fixture: returned ViTAE_Window_NoShift_basic · level recorded 2026-09-24
conv1x1 in 691 papers resnet.py ran · our draft was wrong CC0-1.0 (permissive) here; the stored copy is pointer only · 6ef392f365cc5703 · report fixture: signature in_planes:int/0/int, out_planes:int/0/int, stride:int/0/int · returned Conv2d · level recorded 2026-09-22
conv3x3 in 1,376 papers resnet.py ran · our draft was wrong CC0-1.0 (permissive) here; the stored copy is pointer only · 794a2ed91a5175bd · report fixture: signature in_planes:int/0/int, out_planes:int/0/int, stride:int/0/int, groups:int/0/int, dilation:int/0/int · returned Conv2d · level recorded, date not recorded
get_norm_layer in 111 papers networks.py ran CC0-1.0 (permissive) · 5030170039952082 · report fixture: signature norm_type:str/0/str · returned partial · level recorded 2026-09-24
get_scheduler in 122 papers networks.py ran CC0-1.0 (permissive) · 5fd6b26c18104c25 · report fixture: signature optimizer:Adam/0/?, args:Args/0/? · returned LambdaLR · level recorded 2026-09-24
get_sinusoid_encoding in 24 papers NormalCell.py ran fingerprinted CC0-1.0 (permissive) · 2c3dfd32c766d157 · report fixture: signature n_position:int/0/int, d_hid:int/0/int · returned Tensor · level recorded 2026-09-24
init_net in 47 papers networks.py ran CC0-1.0 (permissive) · 2009bbd28575495d · report fixture: signature net:callable/0/?, init_type:str/0/str, init_gain:float/0/float, gpu_ids:list/1/empty · returned SimpleNet · level recorded 2026-09-24
test_path utils/dataloaders.py ran CC0-1.0 (permissive) · d0814d9bf3797cb1 · report fixture: signature data_dir:str/0/str · returned dict · level recorded 2026-09-24
train_path utils/dataloaders.py ran CC0-1.0 (permissive) · 1203a926bfe91b55 · report fixture: signature data_dir:str/0/str · returned tuple · level recorded 2026-09-24
window_reverse in 296 papers swin.py ran · our draft was wrong CC0-1.0 (permissive) · 61bf152e6a42a184 · report fixture: signature windows:Tensor/4/float, window_size:int/0/int, H:int/0/int, W:int/0/int · returned Tensor · level recorded, date not recorded
yangyucheng000/papercode-2 community (archive-listed) 12 samples, 9 ran (0 instrument-failed) · 3 unverified
Attention in 344 papers
census: 13 implementations from 12 papers, 10 distinct outputs across 4 buckets, one shared input per bucket
STViT-Mindspore-main/models/stvit.py ran Apache-2.0 (permissive) · 3fefa875010c6edf · report fixture: returned Attention · level recorded 2026-09-22
Fold STViT-Mindspore-main/models/stvit.py ran Apache-2.0 (permissive) · a74921de2215fc04 · report fixture: returned Fold · level recorded 2026-09-22
LayerNorm2d in 18 papers STViT-Mindspore-main/models/stvit.py ran Apache-2.0 (permissive) · 48cf1b51f86c6f57 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
Mlp in 102 papers STViT-Mindspore-main/models/stvit.py ran Apache-2.0 (permissive) · 5a8382845c81cc22 · report fixture: returned Mlp · level recorded 2026-09-22
PatchEmbed in 123 papers
census: 9 implementations from 9 papers, 6 distinct outputs on one shared input
STViT-Mindspore-main/models/stvit.py ran fingerprinted Apache-2.0 (permissive) · dedd2d6dd673cc46 · report fixture: returned PatchEmbed · level recorded 2026-09-22
PatchMerging in 31 papers STViT-Mindspore-main/models/stvit.py ran fingerprinted Apache-2.0 (permissive) · e8f2f05db5500fa5 · report fixture: returned PatchMerging · level recorded 2026-09-22
ResDWC STViT-Mindspore-main/models/stvit.py ran Apache-2.0 (permissive) · b07535c6bcc4574f · report fixture: returned ResDWC · level recorded 2026-09-22
StokenAttention STViT-Mindspore-main/models/stvit.py ran Apache-2.0 (permissive) · 20466726ccd0930d · report fixture: returned StokenAttention · level recorded 2026-09-22
Unfold STViT-Mindspore-main/models/stvit.py ran fingerprinted Apache-2.0 (permissive) · 08e6e488af05edb9 · report fixture: returned Unfold · level recorded 2026-09-22
BasicLayer in 56 papers STViT-Mindspore-main/models/stvit.py unverified Apache-2.0 (permissive) · 6ff0493f05a12074 · report
STViT STViT-Mindspore-main/models/stvit.py unverified Apache-2.0 (permissive) · b1693bc0a5bf9e5b · report
StokenAttentionLayer STViT-Mindspore-main/models/stvit.py unverified Apache-2.0 (permissive) · 6db1322f923dc42b · report
ViTAE-Transformer/ViTAE-VSA community 2 samples, 2 ran (2 instrument-failed) · 0 unverified
window_partition in 377 papers
census: 13 implementations from 17 papers, 5 distinct outputs across 3 buckets, one shared input per bucket
Object-Detection/mmdet/models/backbones/vitaev2_vsa_modules/window.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted no licence file found · pointer only · 61c027be5178a479 · report fixture: signature x:Tensor/4/float, window_size:int/0/int · returned Tensor · level recorded, date not recorded
window_reverse in 296 papers Object-Detection/mmdet/models/backbones/vitaev2_vsa_modules/window.py ran · our draft was wrong no licence file found · pointer only · 5e2dbd713ec13547 · report fixture: signature windows:Tensor/4/float, window_size:int/0/int, H:int/0/int, W:int/0/int · returned Tensor · level recorded, date not recorded

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for ViTAEv2 (graph method: ViTAEv2). 14 bricks; individually on a fixture: 11 ran (of which 9 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

  1. STViT entry · level 0 · unverified · b1693bc0a5bf9e5b
  2. BasicLayer model · level 0 · unverified · 6ff0493f05a12074
  3. MemoryEfficientSwish helper · level 1 · ran · dfd0461ebe45d041
  4. PatchEmbed model · level 1 · ran · dedd2d6dd673cc46
  5. PatchMerging model · level 1 · ran · e8f2f05db5500fa5
  6. StokenAttentionLayer helper · level 0 · unverified · 6db1322f923dc42b
  7. SwishImplementation helper · level 1 · ran · 64a011654b82d503
  8. LayerNorm2d helper · level 1 · ran · 48cf1b51f86c6f57
  9. Mlp model · level 1 · ran · 5a8382845c81cc22
  10. ResDWC model · level 1 · ran · b07535c6bcc4574f
  11. StokenAttention model · level 1 · ran · 20466726ccd0930d
  12. Attention model · level 1 · ran · 3fefa875010c6edf
  13. Fold helper · level 1 · ran · a74921de2215fc04
  14. Unfold helper · level 1 · ran · 08e6e488af05edb9

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2202.10108") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Image ClassificationInductive Bias

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Classification ImageNet ViTAE-H + MAE (448) Number of params 644M #45 of 1060 Archive leaderboard report
Image Classification ImageNet ViTAE-H + MAE (448) Top 1 Accuracy 88.5% #45 of 1060 Archive leaderboard report
Image Classification ImageNet ReaL ViTAE-H (MAE, 512) Accuracy 91.2% #2 of 57 Archive leaderboard report
Image Classification ImageNet ReaL ViTAE-H (MAE, 512) Params 644M #2 of 57 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position EncodingsAdamAttentionBPEConvolutionDense ConnectionsDropoutLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformerVision Transformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections